Commits

Commit history

1. Commits on Apr 9, 2026

   • fix missing change of llama_cpp.llama_split_mode.LLAMA_SPLIT_MODE_LAYER,

     JamePeng

     committed
     f22b819View commit details

     Copy full SHA for f22b819

     View code at this pointBrowse repository at this point
2. Commits on Mar 22, 2026

   • refactor: remove legacy static LoRA initialization

     Show description for 87157f3
     JamePeng

     committed
     87157f3View commit details

     Copy full SHA for 87157f3

     View code at this pointBrowse repository at this point
3. Commits on Jan 29, 2026

   • Sync llama : disable Direct IO by default

     JamePeng

     committed
     5bdff0bView commit details

     Copy full SHA for 5bdff0b

     View code at this pointBrowse repository at this point
4. Commits on Jan 8, 2026

   • Update llama.cpp API 20260108

     JamePeng

     committed
     4d0fe24View commit details

     Copy full SHA for 4d0fe24

     View code at this pointBrowse repository at this point
5. Commits on Jan 1, 2026

   • Update llama_context_params and fix the `embeddings` typo

     JamePeng

     committed
     b6bf564View commit details

     Copy full SHA for b6bf564

     View code at this pointBrowse repository at this point
6. Commits on Nov 2, 2025

   • Implement LlamaTrieCache into llama_cache.py: Optimize LlamaCache lookup from O(N) to O(K) using a Trie

     JamePeng

     committed
     2419dc2View commit details

     Copy full SHA for 2419dc2

     View code at this pointBrowse repository at this point
7. Commits on Aug 31, 2025

   • Sync llama: use FA + max. GPU layers by default

     JamePeng

     committed
     2be720dView commit details

     Copy full SHA for 2be720d

     View code at this pointBrowse repository at this point
8. Commits on Aug 26, 2025

   • Sync llama : remove KV cache defragmentation logic

     JamePeng

     committed
     ae46275View commit details

     Copy full SHA for ae46275

     View code at this pointBrowse repository at this point
9. Commits on Jul 17, 2025

   • fix memory_seq_rm crash bug

     JamePeng

     committed
     8d981f0View commit details

     Copy full SHA for 8d981f0

     View code at this pointBrowse repository at this point
10. Commits on Jul 16, 2025

    • Update llama.cpp API 20250717

      JamePeng

      committed
      bc283d3View commit details

      Copy full SHA for bc283d3

      View code at this pointBrowse repository at this point
11. Commits on Jul 13, 2025

    • more logits_all params restored

      JamePeng

      committed
      cdc7348View commit details

      Copy full SHA for cdc7348

      View code at this pointBrowse repository at this point
12. Commits on May 20, 2025

    • Sync kv-cache : add SWA support

      JamePeng

      committed
      177445cView commit details

      Copy full SHA for 177445c

      View code at this pointBrowse repository at this point
13. Commits on May 13, 2025

    • Update llama.cpp API code 20250513

      JamePeng

      committed
      6b3e11cView commit details

      Copy full SHA for 6b3e11c

      View code at this pointBrowse repository at this point
14. Commits on May 9, 2025

    • Sync context : remove logits_all flag and update API

      JamePeng

      committed
      1a1d920View commit details

      Copy full SHA for 1a1d920

      View code at this pointBrowse repository at this point
15. Commits on Sep 20, 2024

    • feat: Add option to configure n_ubatch

      abetlen

      committed
      6c44a3fView commit details

      Copy full SHA for 6c44a3f

      View code at this pointBrowse repository at this point
16. Commits on Jul 9, 2024

    • fix(misc): Format

      abetlen

      committed
      c1ae815View commit details

      Copy full SHA for c1ae815

      View code at this pointBrowse repository at this point
17. Commits on Jul 2, 2024

    • fix(server): Update `embeddings=False` by default. Embeddings should be enabled by default for embedding models.

      abetlen

      committed
      bf5e0bbView commit details

      Copy full SHA for bf5e0bb

      View code at this pointBrowse repository at this point
18. Commits on Jun 4, 2024

    • feat: adding `rpc_servers` parameter to `Llama` class (#1477)

      Show description for d634efc
      

      chraac

      and

      abetlen

      authored
      d634efcView commit details

      Copy full SHA for d634efc

      View code at this pointBrowse repository at this point
19. Commits on May 5, 2024

    • feat(server): Add support for setting root_path. Closes #1420

      abetlen

      committed
      0318702View commit details

      Copy full SHA for 0318702

      View code at this pointBrowse repository at this point
20. Commits on Apr 30, 2024

    • feat: Add option to enable `flash_attn` to Lllama params and ModelSettings

      abetlen

      committed
      22d77eeView commit details

      Copy full SHA for 22d77ee

      View code at this pointBrowse repository at this point
21. Commits on Apr 26, 2024

    • fix: pydantic deprecation warning

      abetlen

      committed
      fcfea66View commit details

      Copy full SHA for fcfea66

      View code at this pointBrowse repository at this point
22. Commits on Apr 23, 2024

    • feat(server): Provide ability to dynamically allocate all threads if desired using `-1` (#1364)

      sean-bailey

      authored
      53ebcc8View commit details

      Copy full SHA for 53ebcc8

      View code at this pointBrowse repository at this point
23. Commits on Apr 17, 2024

    • feat: add `disable_ping_events` flag (#1257)

      Show description for b73c73c
      khimaros

      authored
      b73c73cView commit details

      Copy full SHA for b73c73c

      View code at this pointBrowse repository at this point

    • feat: Use all available CPUs for batch processing (#1345)

      

      ddh0

      authored and

      abetlen

      committed
      c96b2daView commit details

      Copy full SHA for c96b2da

      View code at this pointBrowse repository at this point
24. Commits on Apr 1, 2024

    • feat: add support for KV cache quantization options (#1307)

      Show description for f165048
      

      Limour-dev

      and

      abetlen

      authored
      f165048View commit details

      Copy full SHA for f165048

      View code at this pointBrowse repository at this point
25. Commits on Feb 28, 2024

    • misc: Format

      abetlen

      committed
      727d60cView commit details

      Copy full SHA for 727d60c

      View code at this pointBrowse repository at this point

    • feat: Update llama.cpp

      abetlen

      committed
      0d37ce5View commit details

      Copy full SHA for 0d37ce5

      View code at this pointBrowse repository at this point
26. Commits on Feb 26, 2024

    • feat(server): Add support for pulling models from Huggingface Hub (#1222)

      Show description for 4d574bd
      abetlen

      authored
      4d574bdView commit details

      Copy full SHA for 4d574bd

      View code at this pointBrowse repository at this point

    • fix: remove prematurely commited change

      abetlen

      committed
      dcf38f6View commit details

      Copy full SHA for dcf38f6

      View code at this pointBrowse repository at this point
27. Commits on Feb 25, 2024

    • feat: Update llama.cpp

      abetlen

      committed
      2292af5View commit details

      Copy full SHA for 2292af5

      View code at this pointBrowse repository at this point
28. Commits on Feb 17, 2024

    • feat: Update llama.cpp

      abetlen

      committed
      fdce078View commit details

      Copy full SHA for fdce078

      View code at this pointBrowse repository at this point
29. Commits on Jan 31, 2024

    • Add speculative decoding (#1120)

      Show description for fb762a6
      abetlen

      authored
      fb762a6View commit details

      Copy full SHA for fb762a6

      View code at this pointBrowse repository at this point
30. Commits on Jan 29, 2024

    • Automatically set chat format from gguf (#1110)

      Show description for da003d8
      abetlen

      authored
      da003d8View commit details

      Copy full SHA for da003d8

      View code at this pointBrowse repository at this point
31. Commits on Jan 19, 2024

    • feat: Add ability to load chat format from huggingface autotokenizer or tokenizer_config.json files.

      abetlen

      committed
      b8fc1c7View commit details

      Copy full SHA for b8fc1c7

      View code at this pointBrowse repository at this point
32. Commits on Jan 18, 2024

    • Offload KQV by default

      abetlen

      committed
      48c3b77View commit details

      Copy full SHA for 48c3b77

      View code at this pointBrowse repository at this point